Abstract 179

Submitted by Szymon GLADYSZ

Authors

S. Gladysz (1), E. Ribak (2), B. Femenia Castella (3), L. Jolissaint (4), P. Martinez (1), R. Soummer (5), E. Aller Carpentier (1), M. Kasper (1), J. Christou (6)

Affiliations

(1)European Southern Observatory; (2)Israel Institute of Technology; (3)Instituto de Astrofísica de Canarias; (4)Leiden Observatory; (5)Space Telescope Science Institute; (6)Gemini Observatory

Abstract

Imaging exoplanets requires extreme adaptive optics, coronagraphy and post-processing methods. The capability of a planet imager to detect faint companions will in the end be limited by its ability to remove stellar light. We propose a novel and simple technique to suppress this source of noise.

The method is based on the realization that by obscuring part of the wavefront in the Lyot stop we can change the shape of the corresponding PSF. This obscuration is then moved (rotated or expanded) leading to modulated Airy rings with the speckles pinned to them. This modulation could then be removed from the set of images, leading to better SNR of the possible planets.

To test this approach we wrote a simulation of the E-ELT, taking into account static aberrations in the telescope. We modeled an extreme adaptive optics system, a focal plane mask, and a Lyot stop with orientation or size changing with time. Having a set of simulated star-planet images there are different ways to analyse the data. The methods rely on the fact that the intensity in each pixel of the stellar image is modulated by the Lyot stop reduction. Only a very small part of this intensity stays constant, that of the planet, if it exists at a particular location. With this knowledge we can use the dark-speckle algorithm, speckle discrimination, or techniques proposed for imaging spectroscopy, such as spectral deconvolution. The aperture modulation method requires at least several photons from the companion in every frame, and so for very small and faint exoplanets our technique benefits from the ELT’s huge light-collecting area.

We compared simulated images of multi-planet systems obtained with the use of the classical Lyot coronagraph to images acquired with our device. We found an order of magnitude improvement in terms of SNR of the planets. We plan to test our setup on the High Order Testbench at ESO.